A light field camera relates to an imaging technology that can be used for recording and reproducing a three-dimensional scene by using a sub-lens array. Generally, a sub-lens array is arranged between a main lens and an image sensor such as CCD, so as to record light field information of a three-dimensional scene in different directions by using the sub-lens array on a focusing plane of the sub-lens array.
Different from a two-dimensional image capture manner of a traditional camera, a light field camera can record four-dimensional light field information (such as space and views) of a three-dimensional scene through a single exposure, and support “shooting first and focusing later” (that is, focusing needs not to be performed during shooting). A specific range of a light field image after being shot around a focusing plane during shooting is referred to as a depth range of refocusing point adjustment of the light field image. The depth range of refocusing point adjustment of a light field camera may be captured in advance through calculation according to characteristics and relative positions of elements of the light field camera.
Within the depth range of refocusing point adjustment, a new refocusing point may be determined at a depth interval of a certain step size, so as to acquire a refocused image based on the refocusing point from the light field image. A refocusing accuracy may also be referred to as a refocusing spatial depth resolution, which represents the size of a step value of an interval between two adjacent refocusing points within the depth range of refocusing point adjustment. If the refocusing accuracy is greater, a new refocusing point may be acquired at an interval of a smaller step size within the depth range of refocusing point adjustment; otherwise, if the refocusing accuracy is smaller, a new refocusing point may be acquired at an interval of a greater step size within the depth range of refocusing point adjustment.